© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5331
    Prev
    Next
    INDEX
    Explanations

    terms related to economic concepts and themes of whimsy or capriciousness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    568
    +0.11
    0.5%
    76
    +0.10
    0.5%
    526
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.03
    568
    +0.10
    0.02
    326
    +0.10
    0.03
    Negative Logits
     seq
    -0.70
    ledged
    -0.66
     Leone
    -0.64
     Rabbit
    -0.64
    Footnote
    -0.63
     Sands
    -0.60
     Vega
    -0.60
    regon
    -0.60
     Protestants
    -0.60
     ********************************
    -0.59
    POSITIVE LOGITS
    icon
    1.07
    onom
    0.95
    etically
    0.84
    iflower
    0.81
    ical
    0.81
    stros
    0.79
    orph
    0.79
    otion
    0.78
    agnetic
    0.77
    idate
    0.77
    Activations Density 0.162%

    No Known Activations